Unifying rules for command parsing

[SirZenith]

Currently, there are two different rules for command parsing. One for sub commands like str join, another for simple command like echo.

This make application side difficult to tell the difference between echo this and str join, cause they are both being parsed as sub command.

And this two rules also make command with three or more name segments failed to parse.

• This pull request simplify those two rule into one, which always parse and only parse the first command segment as command name, everything else that follows are arguments. Applications can look into arguments on their own, to recognize long command with more than one name segment.
  • Also this pull request rewrite part of pipeline rule, to allow user add new line before and after |.
  • Add an extra argument to ts test command in ts.nu to allow test case name filtering (also added completion for that arg).

[fdncred]

@SirZenith Thanks for this PR!

@1Kinoti Would you mind reviewing this PR and letting us know what you think?

[1Kinoti]

@SirZenith thanks for this.

a few nitpicks and we'll be good to go

[1Kinoti]

@fdncred this is good to go. thanks again @SirZenith

[SirZenith]

😄

[fdncred]

@fdncred this is good to go. thanks again @SirZenith

Thanks @1Kinoti and @SirZenith. We appreciate the support!!